International Journal of Future Generation Communication and NetworkingVol. 5, No. 2, June, 2012 Database Management System as a Cloud Service Yvette E.Gelogo1 and Sunguk Lee2*1Society of Science andEngineering Research Support,[email protected] 2Research Institute of Industrial Science andTechnologyPohang,Gyeongbuk, Korea [email protected]
re.kr*CorrespondentAuthor: Sunguk Lee* ([email protected]) Abstract A Cloud database management system is a distributed database thatdelivers computing as a service instead of a product. It is the sharing ofresources, software, and information between multiple devices over a networkwhich is mostly the internet.
It is expected that this number will growsignificantly in the future. As a result, there is a growing interest inoutsourcing database management tasks to third parties that can provide thesetasks for much lower cost due to the economy of scale just like putting it intothe cloud. In this paper, we discuss the recent trend in database managementsystem and the possibilities of making it as one of the services offered in thecloud. We also proposed an architecture of database management system in thecloud. Keywords:DBMS,Database Management System, Cloud computing 1.
Introduction In recent years, databaseoutsourcing has become an important component of cloud computing. Due to therapid advancements in a network technology, the cost of transmitting a terabyteof data over long distances has decreased significantly in the past decade. Inaddition, the total cost of data management is five to ten times higher thanthe initial acquisition cost. As a result, there is a growing interest inoutsourcing database management tasks to third parties that can provide thesetasks for much lower cost due to the economy of scale. This new outsourcingmodel has the benefits of reducing the cost for running Database ManagementSystem (DBMS) independently 1.
Cloud computing economics leveraging the powerof multi-tenancy delivers extremely fast shared storage at a dramaticallyreduced cost. Virtualization then compounds these advantages by enabling usersto scale elastically and to pay only for the resources they use. Thecost/performance advantages have decisively shifted in favor of the shared-diskDBMS. It is just a matter of time before the shared-disk DBMS establishesdominance in the cloud. A Cloud database managementsystem (CDBMS) is a distributed database that delivers computing as a serviceinstead of a product. It is the sharing of resources, software, and informationbetween multiply devices over a network which is mostly the internet.
It isexpected that this number will grow significantly in the future. An example ofthis is Software as a Service, or SaaS, which is an application that is deliveredthrough the browser to customers. Cloud applications connect to a database thatis being run on the cloud and have varying degrees of efficiency. Some aremanually configured, some are preconfigured, and some are native. Native clouddatabases are traditionally better equipped and more stable that those that aremodified to adapt to the cloud. 71International Journal ofFuture Generation Communication and Networking Vol. 5, No.
2, June, 2012 2. Background 2.1 Database Management System (DBMS) A databasemanagement system (DBMS) is a software package with computer programs thatcontrol the creation, maintenance, and use of a database.
It allowsorganizations to conveniently develop databases for various applications bydatabase administrators (DBAs) and other specialists. A database is anintegrated collection of data records, files, and other objects. A DBMS allowsdifferent user application programs to concurrently access the same database.DBMSs may use a variety of database models, such as the relational model orobject model, to conveniently describe and support applications. It typicallysupports query languages, which are in fact high-level programming languages,dedicated database languages that considerably simplify writing database applicationprograms.
2.2 Cloud Characteristics One of theoft-cited advantages of cloud computing is its elasticity in the face ofchanging conditions. For example, during seasonal or unexpected spikes indemand for a product retailed by an e-commerce company, or during anexponential growth phase for a social networking Website, additionalcomputational resources can be allocated on the fly to handle the increaseddemand in mere minutes (instead of the many days it can take to procure thespace and capital equipment needed to expand the computational resourcesin-house). Similarly, in this environment, one only pays for what one needs, soincreased resources can be obtained to handle spikes in load and then releasedonce the spike has subsided. However, getting additional computationalresources is not as simple as a magic upgrade to a bigger, more powerfulmachine on the fly; rather, the additional resources are typically obtained byallocating additional server instances to a task 3.
Having DBMS in the cloudwill give advantage in fast and elastic computing. 3. DBMS as a Cloud Service Most DBMS ordatabase management systems are simply software packages that users can acquireto create, maintain or use a database. However, since the introduction of cloudcomputing, DBMS has morphed into an entirely new type of service with its ownunique benefits and task specific advantages. For one thing, any type of cloudservice model will have to employ a dedicated cloud DBMS in order to trulyprovide customers with excellent access to data and databases. Traditional DBMS’sare simply not set up or equipped to deal with the demands of cloud computing.
And of course, if DBMS was deployed as a service as part of a larger packageprovided, it would likely be much more efficient in its duties and thereforecheaper in the long run. The concept of theDBMS has been around since the beginning of commercial computing; such as thenavigational DBMS of the1960’s. Database management systems are one of theoldest integral components of computing, essentially making it possible toscan, retrieve and organize data on hard drives and networks. All DBMS, despitewhether traditional or cloud-based, are essentially communicators that functionas middlemen between the operating system and the database. How is a cloud DBMSdifferent a traditional one? For one thing, cloud-based DBMS are extremelyscalable.
They are able to handle volumes of data and processes that wouldexhaust a typical DBMS. Despite their scalability however, cloud DBMS are stillsomewhat lacking in their ability to scale up to extremely large processes;this is expected to be remedied in the coming months and years however.Currently, the use of cloud DBMS’s are principally used in the testing anddevelopment of new cloud applications and processes. But while a stand-aloneDBMS can be used on a cloud infrastructure; 72International Journal of Future Generation Communication and NetworkingVol. 5, No. 2, June, 2012 most are not designed to take fulladvantage of cloud resources. DBMS as a cloud service-type models seek tocapitalize on the disparity between antiquated DBMS models and their lack offull cloud functionality. Cloud DBMS mayutilize all of these components or may have devised new strategies that combineone or more elements (like combining data structures and the data querylanguage, for example).
Many organizations are exploring the option ofutilizing pre-existing modeling languages as a basis for expansion in a cloudmodel. This strategy ultimately saves on the time spent developing cloud DBMS’sas well as enhances their overall effectiveness, since traditional modelinglanguages are more than adequate for handling data. Despite thebenefits offered by cloud-based DBMS, many people still have apprehensionsabout them. This is most likely due to the various security issues that haveyet to be dealt with. These security issues stem from the fact that cloud DBMSare hard to monitor since they often span across multiple hardware stacksand/or servers. Security becomes a serious issue with cloud DBMS when there’smultiple Virtual Machines (which might be accessing databases via any number ofapplications) that might be able to access a database without being noticed orsetting off any alerts. In this type of situation a malicious person couldpotentially access pertinent data or cause serious harm to the integralstructure of a database, putting the entire system in jeopardy.
There is however aproposed method for dealing with these types of incongruence. An obvioussolution is the deployment of an autonomous network agent, which rigorouslymonitor and defends all activities related to database access. The limitationof this method however, is that a network agent may be unable to handleextremely large and dense volumes of activity / traffic. Arguably, the bestsolution for dealing with security issues is to employ continuous databaseauditing. This involves setting up a system that meticulously records, analyzeand report on all activities regarding database access, especially suspiciousdatabase access. All information regarding these activities is logged andstored in an extremely remote and secure location with alerts being sent out tocloud management (or including any other individuals they might have designatedto receive this information) in the event of a breach. This will provide thosein charge of security with the information necessary to determine who isresponsible, where they are located as well as the specifics of their machine /hardware. While deployment of a dedicated andthorough cloud DBMS hasn’t occurred yet, it is certainly under development.
Theemergence of a comprehensive solution for all cloud service models regardingdatabase management will open the door to a new era of cloud computing. Many of these cloud databases aredesigned to run on a cluster of hundreds to thousands of nodes, and are capableof serving data ranging from hundreds of terabytes to petabytes. Compared withtraditional relational database servers, such cloud databases may offer lessquerying capability and often weaker consistency guarantees, but scale muchbetter by providing built-in support on availability, elasticity, and loadbalancing. On the other hand, datamanagement tools are an important part of relational and analytical datamanagement business since business analysts are often not technically advancedand do not feel comfortable interfacing with low-level database softwaredirectly. These tools typically interface with the database using ODBC or JDBC,so database software that want to work these products must accept SQL queries.
Therefore, a novel technology to combine DBMS capability with Cloud scalescalability is highly desirable. 73International Journal ofFuture Generation Communication and Networking Vol. 5, No. 2, June, 2012 4. Why DBMS in Cloud? Database Management Systems as acloud service are engineered to run as a scalable, elastic service available ona cloud infrastructure. These DBMS are available only as a cloud offering andare not necessarily relational. For example, Microsoft’s SQL Azure is fullyrelational DBMS, while Microsoft’s SQL services, Amazon’s simpleDB and Google’sBig Table are not relational and have different persistence models. Cloud-basedDBMS services are provided in a multi-tenancy environment with elasticresources allocation, for use in simple to complex transactions.
DBMS as acloud service excludes those DBMS that will run on the cloud infrastructure,but are not purpose-built as a cloud service. Most of the currently availableDBMS engines will run on cloud infrastructure, but are not specificallyengineered to take advantage of the cloud. This differentiation is the reasonfor the change in name from “DBMS in the Cloud” to “DBMS as a cloud Service”;running on cloud infrastructure does not define a DBMS as a cloud service 2. All currently available cloudDBMS are relatively new. SQL azure, the only fully relational DBMS available,began full production at the beginning of 2012 and still has some sizelimitations; Microsoft plans to reduce, and eventually lift, theserestrictions. Today, DBMS as a cloud serviceare used primarily for development and testing of applications- where databasesizes are small and issues of security and collocation with multiple users arenot concern.
One big advantages of cloud DBMS is their elasticity: the more youuse, the more you pay; the less you use, the less you pay 2. Initially, cloud DBMSs will havean impact for vendors desiring a less expensive platform for development. Ascloud infrastructure with DBMSs gains maturity especially in scalability,reliability and security, cloud implementations used for short-term projectssuch as small departmental applications and rapid development platforms willshow marked cost reductions compared with implementations within the ITdepartment. This advantages reinforced by the ability to set up a cloud DBMSenvironment without the use of expensive IT personnel. The speed of setup willbe a primary driver to rapid deployment of systems without the usualrequirements and planning necessary for IT projects within the IT department.This will also reduce the necessity for IT to respond to short notice and shortduration projects, reducing overall costs in IT. Data management applicationsare potential candidates for deployment in the cloud. This is because an onpremises enterprise database system typically comes with a large, sometimesprohibitive up-front cost, both in hardware and in software.
For many companies(especially for start-ups and medium-sized businesses), the pay as- you-gocloud computing model, along with having someone else worrying aboutmaintaining the hardware, is very attractive. Due to the ever-increasing needfor more analysis over more data in today’s corporate world, along with anarchitectural match in currently available deployment options, we conclude thatread-mostly analytical data management applications are better suited fordeployment in the cloud than transactional data management applications. Wethus outline a research agenda for large scale data analysis in the cloud,showing why currently available systems are not ideally-suited for clouddeployment, and arguing that there is a need for a newly designed DBMS,architected specifically for cloud computing platforms 3. 74International Journal of Future Generation Communication and NetworkingVol.
5, No. 2, June, 2012 5. DBMS in Cloud Architecture Figure 1. DBMS in the Cloud Architecture Above is a proposed DBMS in CloudArchitecture, first layer is the storage, followed by databases and the upperlayer is application layer. In terms of performance, it provides efficient dataaccess with a better distribution of values for some data. Stores frequentlyused SQL statements in memory, avoiding the need for time-consumingrecompilation at run-time. Produces a detailed report on each step used fordata access, allowing you to accurately implement performance enhancements.
Data is encrypted when stored or backed up, without any need for programming toencrypt and decrypt. 6. Conclusion Database Management Systems as acloud service are engineered to run as a scalable, elastic service available ona cloud infrastructure. Cloud DBMSs will have an impact for vendors desiring aless expensive platform for development. In this paper, we presented the ideaof DBMS in the cloud, the possibilities to be offered as one of the servicesoffered by promising capability of cloud computing, that is to be a DBMS as aService. In this paper we proposed an architecture of DBMS in the cloud. References 1 Buyya R, Broberg J and GoscinskiA, “Cloud computing Principles and Paradigms”, A Jon Wiley & Sons, Inc.Publication, (2011).
2 Feinberg D, “DBMS as a Cloud Service”, (2010), Gartner, Inc. and/or itsAffiliates.3 Abadi D, “Data Management in the Cloud: Limitations and Opportunities”, Bulletin of the IEEE ComputerSociety Technical Committee on Data Engineering, (2009). 4 Kellogg D, “DBMS in the Cloud:Amazon SimpleDB”, http://kellblog.com/2007/12/18/dbms-in-the-cloud-amazon-simpledb/.
75International Journal ofFuture Generation Communication and Networking Vol. 5, No. 2, June, 2012 5 GravelleR, “Should You Move Your MySQL Database to the Cloud?”, http://www.
databasejournal.com/features/mssql/should-you-move-your-mysql-database-to-the-cloud.html. 6 Hsieh M, Chang C, Ho L, Wu J andLiu P, “SQLMR: A Scalable Database Management System for Cloud Computing”, InProceedings of ICPP, (2011), pp.315-324. 7 Hogan M, “Database Virtualization and the Cloud”,ScaleDB Inc., (2009).